Medical Data mining is the search for relationships and patterns within the medical datasets that could provide useful knowledge for effective clinical decisions. The medical data are usually multidimensional,and is presented by a large number of features. The inclusion of irrelevant, redundant and noisy features in the process model can result in poor predictive accuracy and increased computation. The success of the data mining technique is based on the selection of a small set of highly predictive attributes. Feature selection techniques extract informative features from medical databases to facilitate medical decision making in an effective and efficient way. The feature selection methods discussed, both Wrapper and Filter methods, use feature ranking methods to identify the informative and discriminatory features of the medical datasets, which greatly influence the predictive accuracy of the classifiers. There is a significant improvement in the sensitivity and specificity values which shows that the feature selection indeed improves the discriminating ability of the classifiers. The focus is on aggressive dimensionality reduction with an increase in the prediction accuracy.